Complementary Protective MeasuresControl FunctionsGeneralRobotics

Why you should stop using the term ‘Deadman’

The Deadman Control

Do you use the phrase ‘dead­man’ or ‘dead­man switch’ when talk­ing about safety-related con­trols on your machinery? I often run into this when I’m work­ing with cli­ents who use the terms to refer to ‘enabling devices’ – you know, those two or three-pos­i­tion switches that are found on robot teach­ing pendants and in oth­er applic­a­tions to give the oper­at­or a way to stop machinery, even if they have already been injured or killed by the equip­ment. Call­ing these devices a ‘Dead­man Switch’ or even a ‘Live-Man Switch’ as the three-pos­i­tion devices are some­times called, sends entirely the wrong mes­sage to the user as far as I’m con­cerned. The object­ive of our work as machinery safety engin­eers is to pre­vent injur­ies from hap­pen­ing in the first place. Using a device that is designed to determ­ine if the user is dead or uncon­scious means someone screwed up!

A little history

The term ‘dead­man’ comes from a device that was developed in the 1880’s by pion­eer­ing elec­tric­al engin­eer Frank Sprague. Sprague was work­ing on elec­tric trac­tion motor tech­no­logy, using these new machines to power street rail­ways (street­cars) and elec­tric elev­at­ors. The early DC motor con­trols used in both street­cars and elev­at­ors required an oper­at­or. The oper­at­or used a hand con­trol to move the street­car for­ward or back­ward along the track and to con­trol the speed of the car. In elev­at­ors, the oper­at­or used a sim­il­ar hand con­trol to move the elev­at­or car up or down the shaft, to con­trol the speed and to stop at the appro­pri­ate floor.

Westinghouse Streetcar Controller
1920’s era West­ing­house Street­car Con­trol­ler

If the oper­at­or was to doze off or fall uncon­scious, the street­car would simply con­tin­ue on its way until it hit some­thing or derailed, either being a poor option! Elev­at­ors would con­tin­ue until they hit the top or bot­tom of the shaft, again a bad idea. Sprague included a con­trol device in his designs that required the oper­at­or to keep his hand on the con­trol­ler handle and to main­tain pres­sure on the con­trol device in order for power to flow to the motor. This same idea was imple­men­ted in manu­al elev­at­or con­trol handles. These ideas were adop­ted by West­ing­house when they developed the street­car motor con­trol­lers that were used in thou­sands of street­cars run­ning between the 1890’s and the 1930’s.

When dies­el-elec­tric and full elec­tric loco­mot­ives were developed in the 1930’s, the concept of the ‘dead­man’ con­trol was adop­ted from street rail­ways. There is a per­sist­ent myth that these con­trols star­ted with steam loco­mot­ives. In fact, an earli­er ver­sion of this art­icle included that myth – now bus­ted!

A 'deadman' pedal in a locomotive.
A ‘dead­man’ ped­al in a dies­el-elec­tric rail­way loco­mot­ive

With the advent of elec­tric trams, trains, and sub­ways, con­cerns about pos­sib­il­it­ies like heart attacks and oth­er infirm­it­ies res­ult­ing in drivers los­ing con­trol of these machines caused these devices to be integ­rated into these new trans­port­a­tion sys­tems. To learn more about these applic­a­tions, see the Wiki­pe­dia art­icle Dead Man’s Switch.

It’s worth not­ing that the rail­ways now call these devices ‘Driver Safety Devices’ or DSD. See a mod­ern DSD at the Arrow­vale Elec­tron­ics web site.

Elev­at­ors moved from manu­al con­trol to auto­mat­ic con­trol, elim­in­at­ing the need for elev­at­or oper­at­ors and the need for ‘dead­man’ con­trols.

Robots Enter the Picture

Motoman robot pendant enabling device
Moto­man pendant with show­ing enabling device (red arrow)

In the 1980’s, indus­tri­al robots began to appear in the work­place. Acci­dents in these early days drove changes in the design of the con­trol pendants used to ‘teach’ these devices their tasks. Early pendants provided motion con­trol and an emer­gency stop device. Later, the motion con­trols were altered to become ‘hold-to-run’ devices that could jog the selec­ted robot axis at a pre-selec­ted slow-speed, one axis at a time. In the 90’s the ‘enabling device’ was added to the pendant. These two-pos­i­tion switches, still called ‘dead-man switches’, had to be held closed in order for the robot to move under con­trol of the axis hold-to-run con­trols. Acci­dents con­tin­ued to occur. In the mid 90’s the three-pos­i­tion enabling device, some­times called a ‘live-man-switch’, was intro­duced after stud­ies showed that some people would release their grip on the con­trol pendant when struck by the robot, while oth­ers would clench the hand hold­ing the pendant. The new switches are required to be held in the mid pos­i­tion to enable motion. The pic­ture at left shows the back of a mod­ern robot pendant. The black bar in the lower right is the enabling device, loc­ated so that your hand will nat­ur­ally hold the device in the cor­rect pos­i­tion when you hold the pendant in your left hand. Not so good if you are left-handed!

ABB IRB640 Robot Pendant
ABB IRB640 Robot Pendant



Euch­ner ZS Switches

In addi­tion to the pendant enabling devices, addi­tion­al enabling devices are required where more than one work­er is required inside the danger zone of the machine. These devices can be pur­chased sep­ar­ately and added to sys­tems as needed. Depend­ing on the applic­a­tion, you can get these devices with emer­gency stop but­tons and jog but­tons integ­rated into a single unit as shown in the pic­ture of the Euch­ner ZS switches.

Machinery Standards and Definitions

The enabling device is one of those pro­tect­ive meas­ures that can­not be read­ily clas­si­fied as a safe­guard­ing device because they do not pro­act­ively pre­vent injury. Instead, like an emer­gency stop, they may allow a work­er to avert or lim­it the harm that is already occur­ring. That places the enabling device into the ‘com­ple­ment­ary pro­tect­ive meas­ure’ cat­egory.

Let’s take a minute to look at a couple of import­ant defin­i­tions from the machinery stand­ards. At the moment, the best defin­i­tion for a com­ple­ment­ary pro­tect­ive meas­ure comes from the Cana­dian stand­ard, CSA Z432-04. Excerp­ted from CSA Z432-04, § Com­ple­ment­ary Pro­tect­ive Meas­ures:

Pro­tect­ive meas­ures that are neither inher­ently safe design meas­ures, nor safe­guard­ing (imple­ment­a­tion of guards and/or pro­tect­ive devices), nor inform­a­tion for use may have to be imple­men­ted as required by the inten­ded use and the reas­on­ably fore­see­able mis­use of the machine. Such meas­ures shall include, but not be lim­ited to,

a) emer­gency stop;

b) means of res­cue of trapped per­sons; and

c) means of energy isol­a­tion and dis­sip­a­tion.

Let’s also look at the form­al defin­i­tion of an ‘enabling device’ in the same stand­ard:

7.23.3 Enabling devices
An enabling device is an addi­tion­al manu­ally oper­ated 2- or 3-pos­i­tion con­trol device used in con­junc­tion with a start con­trol and which, when con­tinu­ously actu­ated in one pos­i­tion only, allows a machine to func­tion. In any oth­er pos­i­tion, motion is stopped or a start is pre­ven­ted.
Enabling devices shall have the fol­low­ing fea­tures:

a) They shall be con­nec­ted to a Cat­egory 0 or a Cat­egory 1 stop (see NFPA 79).

b) They shall be designed in accord­ance with ergo­nom­ic prin­ciples:

(i) pos­i­tion 1 is the off func­tion of the switch (actu­at­or is not oper­ated);

(ii) pos­i­tion 2 is the enabling func­tion (actu­at­or is oper­ated); and

(iii) pos­i­tion 3 (if used) is the off func­tion of the switch (actu­at­or is not oper­ated past its mid pos­i­tion).

c) Three-pos­i­tion enabling devices shall be designed to require manu­al oper­a­tion in order to reach pos­i­tion 3.

d) When return­ing from pos­i­tion 3 to pos­i­tion 2, the func­tion shall not be enabled.

e) An enabling device shall auto­mat­ic­ally return to its off func­tion when its actu­at­or is not manu­ally held in the enabling pos­i­tion.

Note: Tests have shown that human reac­tion to an emer­gency may be to release an object or to hold on tight­er, thus com­press­ing an enabling device. The ergo­nom­ic issues of sus­tained activ­a­tion should be con­sidered dur­ing design and install­a­tion of the enabling device.


OMRON A4EG Enabling Switches
OMRON A4EG Enabling Switches

Sim­il­ar defin­i­tions exist in the Inter­na­tion­al, European and US stand­ards, although they may not be quite as form­al­ized.


Most enabling devices on their own do noth­ing except PERMIT motion to take place, although the actu­al defin­i­tion of enabling device in CSA Z432-04 actu­ally per­mits the enabling device to cause motion. Absence of the enabling sig­nal pre­vents or stops motion. These devices are then used in con­junc­tion with hold-to-run con­trols on robots and machinery, and with throttle con­trols on trains, street cars, sub­ways and sim­il­ar equip­ment. Note that most stand­ards to not per­mit enabling devices to actu­ally cause motion. This is a unique situ­ation in the Cana­dian stand­ard.

So what’s the big deal?

Using the terms ‘dead-man’ or ‘live-man’ to describe these devices puts the wrong mes­sage out as far as I’m con­cerned. As safety engin­eers and OHS prac­ti­tion­ers, we care about keep­ing work­ers out of danger. This is neither check­ing to see if we have either a ‘dead man’ or a ‘live man’, but rather ensur­ing that the per­son in con­trol of the equip­ment is ‘in con­trol’.  Using a phrase like ‘enabling device’ clearly says what the device does.

In my opin­ion, and  sup­por­ted by the cur­rent Inter­na­tion­al and Cana­dian Stand­ards, these terms must be aban­doned in favour of ‘enabling device’ and the qual­i­fi­ers ‘2-pos­i­tion enabling device’ and ‘3-pos­i­tion enabling device’. These terms are also used in many of the cur­rent machinery safety stand­ards, so using them cor­rectly improves clar­ity in writ­ing and speak­ing. Clar­ity in com­mu­nic­a­tion in safety is too import­ant for prac­ti­tion­ers to per­mit the ongo­ing use of terms that con­vey the wrong mes­sage and do not pro­mote clar­ity of mean­ing. Since clar­ity is often lack­ing when it comes to safety, any­thing we can do to improve our com­mu­nic­a­tions should be high on our pri­or­ity list!

Ed. note: This post was updated on 17-Aug-17. The myth of dead­man con­trols on steam loco­mot­ives was removed and replaced by his­tor­ic­ally veri­fi­able inform­a­tion about the ori­gins of this con­trol.

17 thoughts on “Why you should stop using the term ‘Deadman’

  1. I don’t agree in the slight­est. Euphem­isms have no place in H&S. Call the damn thing what it is instead of being too proud to admit there could be a cock-up that requires its use. Next thing you know we’ll be call­ing it by some silly abbre­vi­ation like PPE (we used to call that safety gear if you can remem­ber that far back).
    A dead­man situ­ation is abso­lutely not the right occa­sion for going look­ing in news­peak dic­tion­ar­ies.
    “Enabling but­ton” espe­cially I find ridicu­lous. An on/off but­ton enables. This is much more spe­cif­ic than that.
    Don’t change a single thing. If it’s already been changed by snow­flake legis­lat­ors, admit the mis­take and revert it.

    1. Dave,

      Thanks for your reply. I can see your point, how­ever, this is not about being polit­ic­ally cor­rect or some oth­er notion involving snow­flakes. 🙂 Enabling devices have one pur­pose in life: to enable haz­ard­ous motion under enhanced safety con­di­tions. This means “enhanced” as com­pared to not hav­ing the enabling device or any oth­er safe­guard­ing. The phrase “enabling device” accur­ately describes the func­tion of the device: it enables haz­ard­ous motion. Using the term “dead­man” indic­ates that you’ve already failed and the work­er is now dead. You just want to stop the machinery mangling the corpse too much before you can extract it for the fam­ily. THAT to me is unac­cept­able ter­min­o­logy.

      In any case, change already happened. The term “enabling device” is firmly embed­ded in all of the rel­ev­ant stand­ards and has been for quite a few years. Root­ing it out will likely be very dif­fi­cult to do, espe­cially to replace it with what is now con­sidered out­dated lan­guage.

      1. We’ll agree to dif­fer then.
        What is the pur­pose of a dead­man switch? It’s not to stop a corpse being mangled. It’s to stop the train run­ning away and gate­crash­ing a level cross­ing or whatever. The oper­at­or is dis­abled so you want the machine to stop because nobody leaves uncon­trolled machinery run­ning. Call it a fail-safe or sign-of-life switch if you prefer, but make it mean the same as its func­tion, which is to stop by default, not to enable.

        I actu­ally stumbled on this blog because I was search­ing for the most com­mon usage: dead­man switch? but­ton? dead man’s handle?
        See I’m trans­lat­ing from French to Eng­lish, and guess what it’s called in French par­lance.

        1. Dave,

          Inter­est­ing. Are you read­ing my blog in French? Since I write in Eng­lish and am a nat­ive Eng­lish speak­er I am inter­ested.

          The usage of the term “dead­man” in the rail sec­tor is dif­fer­ent than in the machinery sec­tor. You are cor­rect about the usage in rail loco­mot­ives and light rail vehicles like sur­face rail sub­ways and street­cars. These are machinery, but they are sub­ject to pre­dom­in­antly manu­al con­trol. They are also gen­er­ally not sub­ject to the types of safe­guard­ing sys­tems that are com­monly used to reduce risk in the machinery sec­tor. By the way, sorry for the ref­er­ence to mangled corpses, but I felt the need to stop the trend of the con­ver­sa­tion. And I have to dis­agree – it’s quite pos­sible to have a per­son entangled in a run­ning machine for an exten­ded peri­od of time. See (WARNING: Graph­ic!) or for examples. Enabling devices might have helped in these cir­cum­stances.

          In the machinery sec­tor, it is very com­mon to have an indi­vidu­al work­er take on a task in a rel­at­ively isol­ated part of a fact­ory. An example might be pro­gram­ming a robot to do a task. The tech­ni­cian is nor­mally inside the danger zone, often hid­den from view, and work­ing on poten­tially leth­al equip­ment. If some­thing goes wrong, the work­er is right in the line of fire and is very likely to be injured or killed. There are no oth­er safe­guards oth­er than a hold-to-run con­trol, and an enabling device. These are com­ple­men­ted by an emer­gency stop but­ton, but hon­estly, few work­ers hurt in these con­di­tions have the thought to press that but­ton, nor do they have suf­fi­cient time. So the enabling devices are designed to switch off power if com­pressed or released, based on stud­ies that showed that some people will grasp some­thing in their hands with all their strength when being injured, while oth­ers will drop it under the same con­di­tions. This is a com­pletely dif­fer­ent scen­ario than hav­ing a loco­mot­ive engin­eer fall asleep, or become uncon­scious while the train is in motion.

          So, just to wrap this up, machinery people call them enabling devices. Rail people call them “dead­man ped­als” or “sign-of-life” devices. I am inter­ested to learn more about your interest in this top­ic, and I would be more than happy to carry on the con­ver­sa­tion via email or Skype as you like.

          1. Doug, I’m not sure what you mean about read­ing your blog in French but I can assure you all the words I see are in Eng­lish.

            I trans­late from French to Eng­lish for a liv­ing and in this case the sub­ject was a tor­pedo hand­ling sys­tem (I don’t think I’m giv­ing any mil­it­ary secrets away say­ing that!). The thing is, although the source lan­guage is French they’ve already gone ahead put Eng­lish labels on the hand held con­trol­ler that houses this but­ton, and the word they’ve put on the eponym­ous label is a great big “DEAD MAN”. But I still had to trans­late the descrip­tion.

            They do use the word-for-word equi­val­ent term “Homme mort” in French.
            They also have a term “signe-de-vie” (sign of life), but that requires pos­it­ive action from the oper­at­or every x seconds to keep the thing run­ning, in which case “ena­bler” would be a val­id if less evoc­at­ive descrip­tion. And I think that’s prob­ably the point I’m mak­ing. To me it seems some­thing much more evoc­at­ive than “ena­bler”, which doesn’t give instant con­text, is called for.

            What, I’m won­der­ing, do you call devices where you have to keep the but­ton pressed in order to keep both hands busy, for example on the tail­gate of a deliv­ery van?

            Any­way, I digress. In the tor­pedo example, since the device is loc­ated on a wired remote con­trol unit spe­cific­ally designed to enable the oper­at­or to get far enough away for an over­all view, they aren’t going to be any­where near the machinery (namely an inclined winch) when using it. The only pur­pose I can really see for it is to stop the tor­pedo run­ning out of con­trol down into the sub­mar­ine if the oper­at­or faints or some­thing, which puts it closer to the run­away train cat­egory I sup­pose.

            That was an inter­est­ing point about more soph­ist­ic­ated devices react­ing to the “grip of pan­ic” as well as releas­ing. I wasn’t aware of that, and obvi­ously that’s a much dif­fer­ent situ­ation from those I was refer­ring to. I’ll bear it in mind if it ever comes my way.

            Thanks for the dis­cus­sion


          2. Dave, sorry, I mis­un­der­stood what you meant when you said you were trans­lat­ing from French into Eng­lish – I thought you meant the blog! ? Oops! Any­way, I can def­in­itely see where your com­ments are com­ing from. In the case of the tor­pedo hoist, “dead­man” might very well be appro­pri­ate, con­sid­er­ing what might hap­pen to a sub­merged boat if a tor­pedo got away from the load­ers while it was still inside the boat! I’ve heard rumours that the Rus­si­an sub­mar­ine Kur­sk may have been lost through just such an acci­dent.

            I agree about the sign-of-life devices. These are very com­mon on rail loco­mot­ives and came about because there were a few cases where an engin­eer put his tool­box down on the “dead­man” ped­al so he didn’t have to main­tain pres­sure on the ped­al. There were pre­dict­able res­ults from those escapades as you might ima­gine. I think what we are run­ning into here is dif­fer­ences in usage based on the sec­tor in which sim­il­ar, but dif­fer­ent, con­trol func­tions are used. When it comes down to this, at least from an engin­eer­ing per­spect­ive, this is a safety con­trol func­tion that has a spe­cif­ic out­come – stop­ping unat­ten­ded or out-of-con­trol haz­ard­ous motion.

            The tail­gate devices you men­tion are usu­ally called “two-hand con­trol devices” because they keep the operator’s hands out of the works. These devices are also used on power presses, press brakes, and many oth­er applic­a­tions. There are some very detailed require­ments for how these pro­tect­ive devices are designed and used. For more on that see ISO 13851.

  2. @dougnix:disqus Appre­ci­ated your work. I google the dead man switch, found your blog. Although Eng­lish is not my first lan­guage how­ever, I found it very inform­at­ive and you are right the dead-man switch shall be change to enabling device or hold to run but­ton.

    1. Ejaz Ahmad, thank you. I am glad to hear you found it help­ful. Did you notice the trans­la­tion wid­get at the top of the side­bar? You should be able to trans­late the art­icle into your home lan­guage. If you try this, please let me know if it works well or not. 🙂

  3. The prob­lem with his ana­lys­is is that, basic­ally, he is look­ing at it from the oper­at­or view, infer­ring that oper­a­tion or inop­er­a­tion of the machine only impacts the oper­at­or, while the ori­gin­al pur­pose and its applic­a­tion in the con­tem­por­ary dis­aster was to pro­tect oth­ers from the machine when the oper­at­or did not show pos­it­ive con­trol (i.e. “dead”). There really are two dif­fer­ent prob­lems, pos­sibly war­rant two dif­fer­ent names, but “dead man” is cer­tainly appro­pri­ate for the prob­lem that ori­gin­ated the term.

    1. Ken29,

      Thanks for your com­ment. I don’t dis­agree entirely – the ori­gin­al pur­pose was to pre­vent a run­away machine from injur­ing oth­ers and des­troy­ing itself. The death or dis­ab­il­ity of the oper­at­or was ancil­lary to pro­tect­ing the machine and oth­ers who might be affected.

      Hav­ing said that, the term “dead­man” is no longer used in tech­nic­al stand­ards or in reg­u­la­tions related to safety. These devices have had oth­er names coined for them, depend­ant on the sec­tor. As I men­tion in the art­icle, the rail­ways now call these sys­tems out as “Driver Safety Devices”, while in auto­ma­tion and robot­ics they are termed “enabling devices”. In my opin­ion, both terms are more descript­ive of the func­tion of the device or sys­tem than the old “dead­man”.

      1. Doug: I cer­tainly agree that “dead­man” by itself isn’t par­tic­u­larly inform­at­ive and “dead­man switch” only slightly bet­ter, but any­thing that ends in “device” is pretty ambigu­ous and prob­ably more suited to stand­ards and legis­la­tion than to pop­u­lar ver­nacu­lar. Not that I have thought of a bet­ter altern­at­ive.

        1. ken29,

          I see your point, how­ever the use of the term device allows for new tech­no­lo­gies that may not be acco­mod­ated by “switch”. If you con­sider some of the older devices that have served this pur­pose, valves and switches have both served this pur­pose. Cur­rent stand­ards reflect the change to 3-pos­i­tion devices that incor­por­ate mul­tiple switches and require the user to hold the device in a mid pos­i­tion. This device is designed to ensure that no what reac­tion a user has when injured, the device will remove power from the machinery. Lan­guage needs to advance as well as tech­no­logy. It’s my opin­ion that this change, now more than 10 years old, reflects that advance­ment.

          1. Again, that’s great for tech­ies, stand­ards, and legis­la­tion, but not use­ful or help­ful for every­day com­mu­nic­a­tion with the pub­lic.

          2. No argu­ment. This is always the chal­lenge. How can we most effect­ively con­vey import­ant ideas to lay people, while avoid­ing jar­gon? At the same time, how can we effect­ively com­mu­nic­ate com­plex ideas suc­cinctly when deal­ing with oth­er pro­fes­sion­als.

            What I try to do with this blog is to de-mys­ti­fy com­plex ideas. My goal is to try to help any­one who is inter­ested in under­stand­ing the key ideas related to safety of machinery. There are a lot of mis­con­cep­tions and con­fu­sion out there – I run into it in my prac­tice all the time. If pro­fes­sion­al people work­ing in this sec­tor are con­fused, I’m sure non-pro­fes­sion­als are too.

            So what’s the altern­at­ive in this case? “Dead­man” has been widely used in the past, although it brings with it some ideas that may not be com­pletely cor­rect. “Enabling Device” says what the thing does, but is not as well under­stood. Do you think a third option might exist?

          3. Doug: This isn’t a com­pletely sat­is­fact­ory, or even sat­is­fy­ing, answer, but it seems to come as close as I’ve been able to devise. How about using “con­trol” instead of device, then adding a word or two (as we usu­ally do) to select the spe­cif­ic con­trol?
            I spent some time Googling “device”, “switch”, and “con­trol” to see how they were rep­res­en­ted in Google (sort of con­tem­por­ary ver­nacu­lar) and con­trol seemed to “win” by some mar­gin. I think we got into this con­ver­sa­tion through the train crash out East and for that applic­a­tion I think a “dead­man con­trol” works fairly well and improves on “dead­man switch”. I saw three-func­tion “grip” used on an indus­tri­al machine and that matched the pic­ture well, but I think con­trol would work as well.
            Good luck! It’s a worthy cause.

          4. Ken,

            I agree, “con­trol” cov­ers it effect­ively and could be used to include any kind of “device” or devices neces­sary for any type of applic­a­tion. Good idea!

            Mod­ern rail­way Driver Safety Devices (DSDs) are more com­plex than the ped­al shown in the pic­ture in the art­icle. That one is pretty old-school. I’ve heard from a few people that the ped­al shown in the art­icle often ended up with a tool­box hold­ing it down, defeat­ing the pur­pose of the con­trol.

            Now they usu­ally require the driver to act­ively do some­thing with­in a cer­tain peri­od of time, like press and release a but­ton, or move a con­trol, etc. Every applic­a­tion has it’s own unique require­ments based on what the oper­at­or is required to do to use the machine.

  4. Good tim­ing! I was just at the Auto­mate Show, work­ing in the Robot Safety Stand­ard booth. I was demon­strat­ing some new robot cap­ab­il­it­ies, then said “I hold the enabling device in the cen­ter “on” pos­i­tion”. One of the observ­ers chimed in “the dead­man switch”. I care­fully explained the his­tory – good thing it was the same as the art­icle 😉 And then I explained that a dead­man switch was developed to detect a dead man and stop the train so that more people do not die. How­ever the enabling is used to keep people from harm, have con­trol of the sys­tem, and ENABLE the equip­ment to oper­ate with­in para­met­ers (reduced speed, reduced torque, lim­ited time dur­a­tion,…).

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.